ScholarGate
Assistent

Sammenlign metoder

Gennemgå dine valgte metoder side om side; rækker, der afviger, er fremhævet.

McDonalds Omega til Computerized Adaptive Test×Item Response Theory (IRT)×
FagområdePsykometriPsykometri
FamilieLatent structureLatent structure
Oprindelsesår1999 (omega); CAT application 2000s–2010s1952–1968
OphavspersonRoderick P. McDonald (omega); CAT-omega application extended by IRT and psychometric reliability researchersFrederic M. Lord (and Allan Birnbaum for the 2PL/3PL models)
TypeReliability coefficient for adaptive testsProbabilistic measurement model
Oprindelig kildeMcDonald, R. P. (1999). Test Theory: A Unified Treatment. Lawrence Erlbaum Associates. ISBN: 978-0805830408Lord, F. M. & Novick, M. R. (1968). Statistical Theories of Mental Test Scores. Addison-Wesley. link ↗
AliasserCAT omega reliability, omega in adaptive testing, hierarchical omega for CAT, CAT composite reliabilityIRT, latent trait theory, item characteristic curve theory, modern test theory
Relaterede55
ResuméMcDonald's omega adapted for computerized adaptive testing (CAT) quantifies the reliability of ability or trait estimates when different examinees answer different subsets of items. Unlike Cronbach's alpha, omega is grounded in a factor model, making it suitable for the heterogeneous item pools and variable test lengths that characterize adaptive administrations.Item response theory models the probability that a respondent answers an item correctly (or endorses it) as a function of the respondent's latent trait level and the item's own statistical properties — difficulty, discrimination, and guessing. Unlike classical test theory, IRT places persons and items on the same scale, yielding measurement that is sample-independent for items and test-independent for persons.
ScholarGateDatasæt
  1. v1
  2. 2 Kilder
  3. PUBLISHED
  1. v1
  2. 2 Kilder
  3. PUBLISHED

Gå til søgning Hent slides

ScholarGateSammenlign metoder: CAT McDonald's Omega · Item Response Theory. Hentet 2026-06-19 fra https://scholargate.app/da/compare